[PD] how to avoid (most/many/some) readsf~ dropouts

Andy Farnell padawan12 at obiwannabe.co.uk
Tue Jul 15 02:01:59 CEST 2008



Greg, you may like to experiment with tweaking the disk caches.
For linux see here:

http://www.linuxdevcenter.com/pub/a/linux/2000/06/29/hdparm.html




On Mon, 14 Jul 2008 17:11:02 -0400
"Dr. Greg Wilder" <gregwilder at orpheusmediaresearch.com> wrote:

> Damian Stewart wrote:
> > Dr. Greg Wilder wrote:
> > 
> >> I've attached the latest version of my "basic" playback (w/fade) patch 
> >> for suggestions/comments...  Unfortunately, your ascii patch didn't 
> >> line up, would you mind posting an example patch that shows your method?
> > 
> > first thing i notice when i open it up is this:
> > 
> > on the left hand side, you've got a 'random 50' and a '+ 6', going into 
> > the delay. this means the minimum delay will be 6ms, which i doubt is 
> > long enough.
> > 
> > why the random? why not just hardcode 50ms in there? or at least go 
> > 'random 30' and '+ 40' or something.
> > 
> > hth,
> > d
> > 
> 
> Indeed.  Sorry about that -- I pulled the example from a larger 
> abstraction which added another 2000 to the delay amount, but the +2000 
> didn't get copied.
> 
> At any rate, the "randomized" load time is there to prevent the patch 
> triggering multiple files at exactly the "same" moment.  My 
> understanding is that PD's execution order doesn't allow simultaneous 
> events to collide, however I have found I get better results when I 
> leave a few extra milliseconds between "simultaneous" triggers.  Perhaps 
> it would be smarter to use a hard coded [t b b b]?
> 
> Also, I should point out that the dropouts I experienced happened at 
> different points during playback.  Sometimes near the beginning, other 
> times 50 seconds in...
> 
> G
> 
> 
> _______________________________________________
> Pd-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list


-- 
Use the source




More information about the Pd-list mailing list